Views
No views yet
pvc means the pvc material style but not needed always.figma is the figure style that has joints, and more tend to be product thumbnail images. To get joints, please emphasize figma by 1.2 ~ 1.5 or use doll joints tag.nendoroid means the style of chibi figures. Use with chibi to get better chibis.(worst quality, low quality, bad aesthetic, oldest), bad anatomy, bad hands, text, cropped, signature, blurry
masterpiece, best quality, best aesthetic, 1girl, red hair, cat ears, closed eyes, closed mouth, expressionless, maid costume, apron, lolita dress, frills, lying on back, sheets, red flowers, from above
Negative prompt: (worst quality, low quality, bad aesthetic, oldest), bad anatomy, bad hands, text, error, cropped, jpeg artifacts, signature, watermark, username, blurry
Steps: 18, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 1166304922, Size: 512x768, Model hash: 0866b17d46, Model: pvc-v3-fp16, Denoising strength: 0.7, Clip skip: 2, Hires upscale: 1.5, Hires steps: 12, Hires upscaler: Latent| Parameter | Value |
|---|---|
| Service | Runpod |
| GPU | A4500 |
| Script | kohya-ss/sd-scripts |
| Cost | about $10 |
| Hours | about 30 hours |
| Dataset | about 18000 images from p1atdev/pvc |
| Resolution | 768 |
| Epochs | 13 |
| Optimizer | Lion |
| LR | 1e-7 |
| Scheduler | cosine_with_restarts |
| Train Batch Size | 12 |
1pip install diffusers transformers accelerate scipy safetensors
2pip install xformers1import torch
2from diffusers import StableDiffusionPipeline
3model_id = "p1atdev/pvc-v4"
4pipe = StableDiffusionPipeline.from_pretrained(
5 model_id,
6 torch_dtype=torch.float16,
7)
8pipe = pipe.to("cuda")
9pipe.enable_attention_slicing()
10pipe.enable_xformers_memory_efficient_attention() # required
11prompt = "pvc, masterpiece, best quality, exceptional, 1girl, cat ears, red hair, long hair, hairpin, swept bangs, yellow eyes, black jacket, white shirt, blue tie, white gloves, hand up, upper body, looking at viewer, buildings"
12negative_prompt = "nsfw, nude, worst quality, low quality, bad aesthetic, oldest, bad anatomy"
13image = pipe(
14 prompt,
15 negative_prompt=negative_prompt,
16 guidance_scale=7.0,
17 num_inference_steps=20
18).images[0]
19# save image
20image.save("pvc_figure.png")
21# or just display it
22# display(image)1import torch
2from diffusers import DiffusionPipeline
3model_id = "p1atdev/pvc-v4"
4pipe = DiffusionPipeline.from_pretrained(
5 model_id,
6 torch_dtype=torch.float16,
7 custom_pipeline="lpw_stable_diffusion"
8)
9pipe = pipe.to("cuda")
10pipe.enable_attention_slicing()
11pipe.enable_xformers_memory_efficient_attention() # required
12prompt = """
13pvc, anime, masterpiece, best quality, exceptional,
141girl, bangs, bare shoulders, beret, black hair, black shorts, blue hair, bracelet, breasts, buttons,
15colored inner hair, double-breasted, eyewear removed, green headwear, green jacket, grey eyes, grey sky,
16hat, jacket, jewelry, long hair, looking at viewer, multicolored hair, neck ring, o-ring, off shoulder, rain,
17round eyewear, shorts, sidelocks, small breasts, solo, sunglasses, wavy hair, wet, zipper
18""" # long prompt
19negative_prompt = "nsfw, nude, worst quality, low quality, bad aesthetic, oldest, bad anatomy"
20image = pipe(
21 prompt,
22 negative_prompt=negative_prompt,
23 guidance_scale=7.0,
24 num_inference_steps=20
25).images[0]
26display(image)